I think there should be a warning message if JavaScript is turned off, because you can't run a survey without activated JS at all.

We can run a survey without javascript .....
And WE have too ...

How can that work? There is JS used for submitting the buttons and what about conditions? JS is used for on page validation.

Default behaviour : move = movenext.

For condition : js + PHP.

Try it : it work. Best is to use :

• First question
• Second question, mandatory but hidden by condition on the first question

Have other bug/feature request for condition :
http://bugs.limesurvey.org/view.php?id=7959

Question:

IE6 don't manage good button. It can be reviewed in javascript . But for PHP: what to do ?
If $move!="" and !(in_array($move,$aAcceptedMove)){
$move='movenext'/'movesubmit";
}

Hard to find the best way ...

Think it's better to have accessibility for all button without javascript except for IE6. But IE6 need default value here. Don't want to add an input hidden just for IE6 ....

Add the relationship because can not do test with IE6 with no js in welcome page.

Fix committed to 2.05 branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=12737

Partial : need save/load
Partial put some js for IE6, but no tested

Fix committed to 2.05 branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=12739

Problem for save .... Need rewrite ....

Put some comment here.

• Actual 2.05 system work good on all browser, this can be added for Save button except for one think.

If we have only one input:text question, if user click on "ENTER" without javbascript : it's take the first button (or first input submit).

IN navigator, i add a button move value="next" at start of navogator. This fix problem, a user "ENTER" think he go next.
But Save (or clear all) can be before navigator(prev/next) then default behaviour are "Save".

An idea can be:

• Adding a display:none button at start of navigator.pstpl (think replacement helper can knwo if we are at navigator.pstpl).
• This button can have value="default": default behaviour can be : movenext OR dontmove and send alert (maybe in plugin for 2.06).

It's OK like that ?
Assign it to me after reading and advice.

We do not support IE6 anymore so there is no special IE6 workaround needed.

I really don't understand why they did not find a solution for this in HTML5.
I think the invisible navigator button is the best option although I still hate it ;).

Yes,

After reflexion, think best is to have:
I replacement:
call navigator.pstp do this:
<button name="move" value="default" style="display:none;"></button>
And after put the content of navigator.pstpl.
Then we are sur we have allways value default for move result.

Like i say : the only problem is for "only one text question" and send ENTER.
Some of our user use LS like "Input Assistant"

Fix committed to 2.05 branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=13037

Just fixed a couple of broken paths Denis :)

Fix committed to 2.05 branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=13048

Please nicola_86 : try with the last 2.05 version (or wait for next 2.05 beta release to test).

2.05RC2 released

See https://github.com/LimeSurvey/LimeSurvey/commit/fd5ca4333388f092d69c0133f3a46f323c330d00

We do not support survey taking without JavaScript anymore.

I know for some situation.

But here : we have a working solution, the new solution just need to use a modal in replace of old confirm : i's a "decoration solution".

IE11 : Settings--> Internet Options --> Security --> Custom Level --> "Allow scripting of Microsoft web browser control"

Denis: What is the intention of posting this option of IE? It's not about Javascript support itself, it's about use the IE inside other windows applications.

The issue relating with this option is not confirmed from others.
https://www.limesurvey.org/de/foren/installation-a-update-issues/104844-unable-to-select-multiple-choice-answers
It still looks like a hardened IT environment, where certain features are deactivated. Even though this still might be a bug inside Limesurvey I see no indication that javascript is the issue here.

I am using NoScript everywhere. So I am in favour of keeping a non javascript survey mode in Limesurvey. The question is if the userbase LimeSurvey GmbH is aiming at is the same as the userbase of longtime partners and external codeproviders. And if not, what conclusion can be drawn from that.

I have said it in the past that we will not put any efforts into no-js compatibility.

If someone wants to put time into it, fine, but we cannot guarantee any longterm availablity for the solution. We are not deliberately damaging such a solution but it will also not be considered if there are any changes.